IMPORTING LIBRARIES TO PERFORM EDA

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import matplotlib.gridspec as gridspec
from matplotlib import cm

df1 = pd.read_csv("C:/Users/Ramasish Chatterjee/Downloads/airline_passenger_satisfaction.csv")  ##to read the csv file saved
df1

	Unnamed: 0	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	...	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	arrival_delay_in_minutes	satisfaction
0	0	Male	Loyal Customer	13	Personal Travel	Eco Plus	460	3	4	3	...	5	4	3	4	4	5	5	25	18.0	neutral or dissatisfied
1	1	Male	disloyal Customer	25	Business travel	Business	235	3	2	3	...	1	1	5	3	1	4	1	1	6.0	neutral or dissatisfied
2	2	Female	Loyal Customer	26	Business travel	Business	1142	2	2	2	...	5	4	3	4	4	4	5	0	0.0	satisfied
3	3	Female	Loyal Customer	25	Business travel	Business	562	2	5	5	...	2	2	5	3	1	4	2	11	9.0	neutral or dissatisfied
4	4	Male	Loyal Customer	61	Business travel	Business	214	3	3	3	...	3	3	4	4	3	3	3	0	0.0	satisfied
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
129875	129875	Male	disloyal Customer	34	Business travel	Business	526	3	3	3	...	4	3	2	4	4	5	4	0	0.0	neutral or dissatisfied
129876	129876	Male	Loyal Customer	23	Business travel	Business	646	4	4	4	...	4	4	5	5	5	5	4	0	0.0	satisfied
129877	129877	Female	Loyal Customer	17	Personal Travel	Eco	828	2	5	1	...	2	4	3	4	5	4	2	0	0.0	neutral or dissatisfied
129878	129878	Male	Loyal Customer	14	Business travel	Business	1127	3	3	3	...	4	3	2	5	4	5	4	0	0.0	satisfied
129879	129879	Female	Loyal Customer	42	Personal Travel	Eco	264	2	5	2	...	1	1	2	1	1	1	1	0	0.0	neutral or dissatisfied

129880 rows × 24 columns

df=df1

df.head() #to get top 06 values

	Unnamed: 0	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	...	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	arrival_delay_in_minutes	satisfaction
0	0	Male	Loyal Customer	13	Personal Travel	Eco Plus	460	3	4	3	...	5	4	3	4	4	5	5	25	18.0	neutral or dissatisfied
1	1	Male	disloyal Customer	25	Business travel	Business	235	3	2	3	...	1	1	5	3	1	4	1	1	6.0	neutral or dissatisfied
2	2	Female	Loyal Customer	26	Business travel	Business	1142	2	2	2	...	5	4	3	4	4	4	5	0	0.0	satisfied
3	3	Female	Loyal Customer	25	Business travel	Business	562	2	5	5	...	2	2	5	3	1	4	2	11	9.0	neutral or dissatisfied
4	4	Male	Loyal Customer	61	Business travel	Business	214	3	3	3	...	3	3	4	4	3	3	3	0	0.0	satisfied

5 rows × 24 columns

df.tail()

	Unnamed: 0	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	...	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	arrival_delay_in_minutes	satisfaction
129875	129875	Male	disloyal Customer	34	Business travel	Business	526	3	3	3	...	4	3	2	4	4	5	4	0	0.0	neutral or dissatisfied
129876	129876	Male	Loyal Customer	23	Business travel	Business	646	4	4	4	...	4	4	5	5	5	5	4	0	0.0	satisfied
129877	129877	Female	Loyal Customer	17	Personal Travel	Eco	828	2	5	1	...	2	4	3	4	5	4	2	0	0.0	neutral or dissatisfied
129878	129878	Male	Loyal Customer	14	Business travel	Business	1127	3	3	3	...	4	3	2	5	4	5	4	0	0.0	satisfied
129879	129879	Female	Loyal Customer	42	Personal Travel	Eco	264	2	5	2	...	1	1	2	1	1	1	1	0	0.0	neutral or dissatisfied

5 rows × 24 columns

df = df.drop(columns=['Unnamed: 0'], axis=1)
df

	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	gate_location	...	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	arrival_delay_in_minutes	satisfaction
0	Male	Loyal Customer	13	Personal Travel	Eco Plus	460	3	4	3	1	...	5	4	3	4	4	5	5	25	18.0	neutral or dissatisfied
1	Male	disloyal Customer	25	Business travel	Business	235	3	2	3	3	...	1	1	5	3	1	4	1	1	6.0	neutral or dissatisfied
2	Female	Loyal Customer	26	Business travel	Business	1142	2	2	2	2	...	5	4	3	4	4	4	5	0	0.0	satisfied
3	Female	Loyal Customer	25	Business travel	Business	562	2	5	5	5	...	2	2	5	3	1	4	2	11	9.0	neutral or dissatisfied
4	Male	Loyal Customer	61	Business travel	Business	214	3	3	3	3	...	3	3	4	4	3	3	3	0	0.0	satisfied
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
129875	Male	disloyal Customer	34	Business travel	Business	526	3	3	3	1	...	4	3	2	4	4	5	4	0	0.0	neutral or dissatisfied
129876	Male	Loyal Customer	23	Business travel	Business	646	4	4	4	4	...	4	4	5	5	5	5	4	0	0.0	satisfied
129877	Female	Loyal Customer	17	Personal Travel	Eco	828	2	5	1	5	...	2	4	3	4	5	4	2	0	0.0	neutral or dissatisfied
129878	Male	Loyal Customer	14	Business travel	Business	1127	3	3	3	3	...	4	3	2	5	4	5	4	0	0.0	satisfied
129879	Female	Loyal Customer	42	Personal Travel	Eco	264	2	5	2	5	...	1	1	2	1	1	1	1	0	0.0	neutral or dissatisfied

129880 rows × 23 columns

df.shape

(129880, 23)

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 129880 entries, 0 to 129879
Data columns (total 23 columns):
 #   Column                             Non-Null Count   Dtype  
---  ------                             --------------   -----  
 0   Gender                             129880 non-null  object 
 1   customer_type                      129880 non-null  object 
 2   age                                129880 non-null  int64  
 3   type_of_travel                     129880 non-null  object 
 4   customer_class                     129880 non-null  object 
 5   flight_distance                    129880 non-null  int64  
 6   inflight_wifi_service              129880 non-null  int64  
 7   departure_arrival_time_convenient  129880 non-null  int64  
 8   ease_of_online_booking             129880 non-null  int64  
 9   gate_location                      129880 non-null  int64  
 10  food_and_drink                     129880 non-null  int64  
 11  online_boarding                    129880 non-null  int64  
 12  seat_comfort                       129880 non-null  int64  
 13  inflight_entertainment             129880 non-null  int64  
 14  onboard_service                    129880 non-null  int64  
 15  leg_room_service                   129880 non-null  int64  
 16  baggage_handling                   129880 non-null  int64  
 17  checkin_service                    129880 non-null  int64  
 18  inflight_service                   129880 non-null  int64  
 19  cleanliness                        129880 non-null  int64  
 20  departure_delay_in_minutes         129880 non-null  int64  
 21  arrival_delay_in_minutes           129487 non-null  float64
 22  satisfaction                       129880 non-null  object 
dtypes: float64(1), int64(17), object(5)
memory usage: 22.8+ MB

DATA CLEANING

df.describe(include='all')

	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	gate_location	...	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	arrival_delay_in_minutes	satisfaction
count	129880	129880	129880.000000	129880	129880	129880.000000	129880.000000	129880.000000	129880.000000	129880.000000	...	129880.000000	129880.000000	129880.000000	129880.000000	129880.000000	129880.000000	129880.000000	129880.000000	129487.000000	129880
unique	2	2	NaN	2	3	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	2
top	Female	Loyal Customer	NaN	Business travel	Business	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	neutral or dissatisfied
freq	65899	106100	NaN	89693	62160	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	73452
mean	NaN	NaN	39.427957	NaN	NaN	1190.316392	2.728696	3.057599	2.756876	2.976925	...	3.358077	3.383023	3.350878	3.632114	3.306267	3.642193	3.286326	14.713713	15.091129	NaN
std	NaN	NaN	15.119360	NaN	NaN	997.452477	1.329340	1.526741	1.401740	1.278520	...	1.334049	1.287099	1.316252	1.180025	1.266185	1.176669	1.313682	38.071126	38.465650	NaN
min	NaN	NaN	7.000000	NaN	NaN	31.000000	0.000000	0.000000	0.000000	0.000000	...	0.000000	0.000000	0.000000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000	NaN
25%	NaN	NaN	27.000000	NaN	NaN	414.000000	2.000000	2.000000	2.000000	2.000000	...	2.000000	2.000000	2.000000	3.000000	3.000000	3.000000	2.000000	0.000000	0.000000	NaN
50%	NaN	NaN	40.000000	NaN	NaN	844.000000	3.000000	3.000000	3.000000	3.000000	...	4.000000	4.000000	4.000000	4.000000	3.000000	4.000000	3.000000	0.000000	0.000000	NaN
75%	NaN	NaN	51.000000	NaN	NaN	1744.000000	4.000000	4.000000	4.000000	4.000000	...	4.000000	4.000000	4.000000	5.000000	4.000000	5.000000	4.000000	12.000000	13.000000	NaN
max	NaN	NaN	85.000000	NaN	NaN	4983.000000	5.000000	5.000000	5.000000	5.000000	...	5.000000	5.000000	5.000000	5.000000	5.000000	5.000000	5.000000	1592.000000	1584.000000	NaN

11 rows × 23 columns

df.duplicated().any()

np.False_

df.isnull().sum()

Gender                                 0
customer_type                          0
age                                    0
type_of_travel                         0
customer_class                         0
flight_distance                        0
inflight_wifi_service                  0
departure_arrival_time_convenient      0
ease_of_online_booking                 0
gate_location                          0
food_and_drink                         0
online_boarding                        0
seat_comfort                           0
inflight_entertainment                 0
onboard_service                        0
leg_room_service                       0
baggage_handling                       0
checkin_service                        0
inflight_service                       0
cleanliness                            0
departure_delay_in_minutes             0
arrival_delay_in_minutes             393
satisfaction                           0
dtype: int64

sns.displot(df['arrival_delay_in_minutes'].dropna(),kde=False,bins=50)

<seaborn.axisgrid.FacetGrid at 0x1742a90d0d0>

The above histogram shows that the distribution of arrival_delay_in_minutes is skewed therefore median imputation is a good choice.

df['arrival_delay_in_minutes'] = df['arrival_delay_in_minutes'].fillna(df['arrival_delay_in_minutes'].median())

df.isnull().sum

<bound method DataFrame.sum of         Gender  customer_type    age  type_of_travel  customer_class  \
0        False          False  False           False           False   
1        False          False  False           False           False   
2        False          False  False           False           False   
3        False          False  False           False           False   
4        False          False  False           False           False   
...        ...            ...    ...             ...             ...   
129875   False          False  False           False           False   
129876   False          False  False           False           False   
129877   False          False  False           False           False   
129878   False          False  False           False           False   
129879   False          False  False           False           False   

        flight_distance  inflight_wifi_service  \
0                 False                  False   
1                 False                  False   
2                 False                  False   
3                 False                  False   
4                 False                  False   
...                 ...                    ...   
129875            False                  False   
129876            False                  False   
129877            False                  False   
129878            False                  False   
129879            False                  False   

        departure_arrival_time_convenient  ease_of_online_booking  \
0                                   False                   False   
1                                   False                   False   
2                                   False                   False   
3                                   False                   False   
4                                   False                   False   
...                                   ...                     ...   
129875                              False                   False   
129876                              False                   False   
129877                              False                   False   
129878                              False                   False   
129879                              False                   False   

        gate_location  ...  inflight_entertainment  onboard_service  \
0               False  ...                   False            False   
1               False  ...                   False            False   
2               False  ...                   False            False   
3               False  ...                   False            False   
4               False  ...                   False            False   
...               ...  ...                     ...              ...   
129875          False  ...                   False            False   
129876          False  ...                   False            False   
129877          False  ...                   False            False   
129878          False  ...                   False            False   
129879          False  ...                   False            False   

        leg_room_service  baggage_handling  checkin_service  inflight_service  \
0                  False             False            False             False   
1                  False             False            False             False   
2                  False             False            False             False   
3                  False             False            False             False   
4                  False             False            False             False   
...                  ...               ...              ...               ...   
129875             False             False            False             False   
129876             False             False            False             False   
129877             False             False            False             False   
129878             False             False            False             False   
129879             False             False            False             False   

        cleanliness  departure_delay_in_minutes  arrival_delay_in_minutes  \
0             False                       False                     False   
1             False                       False                     False   
2             False                       False                     False   
3             False                       False                     False   
4             False                       False                     False   
...             ...                         ...                       ...   
129875        False                       False                     False   
129876        False                       False                     False   
129877        False                       False                     False   
129878        False                       False                     False   
129879        False                       False                     False   

        satisfaction  
0              False  
1              False  
2              False  
3              False  
4              False  
...              ...  
129875         False  
129876         False  
129877         False  
129878         False  
129879         False  

[129880 rows x 23 columns]>

VISUAL DATA EXPLORATORY

# Ensure the column names exist in your DataFrame
print(df.columns)

# selection of columns
numerical = df.drop(columns=["Gender", "customer_type", "type_of_travel", "customer_class", "satisfaction"])
categorical = df[["Gender", "customer_type", "type_of_travel", "customer_class", "satisfaction"]]

Index(['Gender', 'customer_type', 'age', 'type_of_travel', 'customer_class',
       'flight_distance', 'inflight_wifi_service',
       'departure_arrival_time_convenient', 'ease_of_online_booking',
       'gate_location', 'food_and_drink', 'online_boarding', 'seat_comfort',
       'inflight_entertainment', 'onboard_service', 'leg_room_service',
       'baggage_handling', 'checkin_service', 'inflight_service',
       'cleanliness', 'departure_delay_in_minutes', 'arrival_delay_in_minutes',
       'satisfaction'],
      dtype='object')

The dataset has slightly more number of female passengers than male passengers.

sns.countplot(x='customer_type', data=categorical)

<Axes: xlabel='customer_type', ylabel='count'>

sns.countplot(x='customer_type', data=categorical)

<Axes: xlabel='customer_type', ylabel='count'>

The proportion of loyal customer are much more than disloyal customer.

sns.countplot(x='type_of_travel', data=categorical)

<Axes: xlabel='type_of_travel', ylabel='count'>

More number of passengers are going for business travel than personal travel.

sns.countplot(x='customer_class', data=categorical)

<Axes: xlabel='customer_class', ylabel='count'>

df['customer_class'].value_counts()

customer_class
Business    62160
Eco         58309
Eco Plus     9411
Name: count, dtype: int64

There are 48%, 7% and 45% customers travelling through business, Economy plus and Economic class respectively.

sns.countplot(x='satisfaction', data=categorical)

<Axes: xlabel='satisfaction', ylabel='count'>

The proportion of satisfied people are less than the neutral or dissatisfied customers.

for x in numerical:
  sns.distplot(numerical, kde=True)
  plt.show() 

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3292991569.py:2: UserWarning: 

`distplot` is a deprecated function and will be removed in seaborn v0.14.0.

Please adapt your code to use either `displot` (a figure-level function with
similar flexibility) or `histplot` (an axes-level function for histograms).

For a guide to updating your code to use the new functions, please see
https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751

  sns.distplot(numerical, kde=True)

sns.countplot(x='inflight_wifi_service',data=df)

<Axes: xlabel='inflight_wifi_service', ylabel='count'>

df['inflight_wifi_service'].value_counts()

inflight_wifi_service
2    32320
3    32185
4    24775
1    22328
5    14356
0     3916
Name: count, dtype: int64

The average rating is 2 or 3 with maximum people have given rating 2 to inflight_wifi_service. This implies inflight_wifi_service has to be improved.

sns.countplot(x='departure_arrival_time_convenient',data=df)

<Axes: xlabel='departure_arrival_time_convenient', ylabel='count'>

df['departure_arrival_time_convenient'].value_counts()

departure_arrival_time_convenient
4    31880
5    27998
3    22378
2    21534
1    19409
0     6681
Name: count, dtype: int64

Maximum number of customers have given rating 4 to departure_arrival_time_convenient. This implies people are satisfied with departure_arrival_time.

sns.countplot(x='ease_of_online_booking',data=df)

<Axes: xlabel='ease_of_online_booking', ylabel='count'>

df['ease_of_online_booking'].value_counts()

ease_of_online_booking
3    30393
2    30051
4    24444
1    21886
5    17424
0     5682
Name: count, dtype: int64

The average rating is 2 or 3 with maximum people have given rating 3 to ease_of_online_booking. This implies quite decent figures.

sns.countplot(x='gate_location',data=df)

<Axes: xlabel='gate_location', ylabel='count'>

df['gate_location'].value_counts()

gate_location
3    35717
4    30466
2    24296
1    21991
5    17409
0        1
Name: count, dtype: int64

The average rating is 3 or 4 with maximum people have given rating 3 to gate_location. This implies people are satisfied with gate_location.

sns.countplot(x='food_and_drink',data=df)

<Axes: xlabel='food_and_drink', ylabel='count'>

df['food_and_drink'].value_counts()

food_and_drink
4    30563
5    27957
3    27794
2    27383
1    16051
0      132
Name: count, dtype: int64

More number of people have given 4 & 5 ratings, However ratings are distributed across 2,3,5 almost equaually with 4 getting the highest count. The ratings are quite varying.

sns.countplot(x='online_boarding',data=df)

<Axes: xlabel='online_boarding', ylabel='count'>

df['online_boarding'].value_counts()

online_boarding
4    38468
3    27117
5    26020
2    21934
1    13261
0     3080
Name: count, dtype: int64

Maximum number of customers have given rating 4 to online_boarding which is quite good.

sns.countplot(x='seat_comfort',data=df)

<Axes: xlabel='seat_comfort', ylabel='count'>

df['seat_comfort'].value_counts()

seat_comfort
4    39756
5    33158
3    23328
2    18529
1    15108
0        1
Name: count, dtype: int64

More number of people have given rating 4 to seat_comfort followed by 5. Maximum number of people are comfortable with seat.

sns.countplot(x='inflight_entertainment',data=df)

<Axes: xlabel='inflight_entertainment', ylabel='count'>

df['inflight_entertainment'].value_counts()

inflight_entertainment
4    36791
5    31544
3    23884
2    21968
1    15675
0       18
Name: count, dtype: int64

More number of people have given rating 4 to inflight_entertainment followed by 5. Maximum number of people are satisfied with inflight_entertainment.

sns.countplot(x='onboard_service',data=df)

<Axes: xlabel='onboard_service', ylabel='count'>

df['onboard_service'].value_counts()

onboard_service
4    38703
5    29492
3    28542
2    18351
1    14787
0        5
Name: count, dtype: int64

More number of people have given rating 4 to onbord_service followed by 5. Maximum number of people are satisfied with onboard_service.

sns.countplot(x='leg_room_service',data=df)

<Axes: xlabel='leg_room_service', ylabel='count'>

df['leg_room_service'].value_counts()

leg_room_service
4    35886
5    30905
3    25056
2    24540
1    12895
0      598
Name: count, dtype: int64

More number of people have given rating 4 to leg_room_service followed by 5. Maximum number of people are satisfied with leg_room_service.

sns.countplot(x='baggage_handling',data=df)

<Axes: xlabel='baggage_handling', ylabel='count'>

df['baggage_handling'].value_counts()

baggage_handling
4    46761
5    33878
3    25851
2    14362
1     9028
Name: count, dtype: int64

More number of people have given rating 4 to baggage_handling followed by 5. Maximum number of people are satisfied with baggage_handling.

sns.countplot(x='checkin_service',data=df)

<Axes: xlabel='checkin_service', ylabel='count'>

df['checkin_service'].value_counts()

checkin_service
4    36333
3    35453
5    25883
1    16108
2    16102
0        1
Name: count, dtype: int64

More number of people have given rating 4 to checkin_service followed by 3. Maximum number of people are quite satisfied with checkin_service, However It can be slightly improved.

sns.countplot(x='inflight_service',data=df)

<Axes: xlabel='inflight_service', ylabel='count'>

df['inflight_service'].value_counts()

inflight_service
4    47323
5    34066
3    25316
2    14308
1     8862
0        5
Name: count, dtype: int64

Maximum number of people have given rating 4 to inflight_service followed by 5. Maximum number of people are satisfied with inflight_service.

sns.countplot(x='cleanliness',data=df)

<Axes: xlabel='cleanliness', ylabel='count'>

df['cleanliness'].value_counts()

cleanliness
4    33969
3    30639
5    28416
2    20113
1    16729
0       14
Name: count, dtype: int64

More number of people have given rating 4 to cleanliness followed by 3. Maximum number of people are quite satisfied with cleanliness.

sns.countplot(x='Gender',data=df,hue='satisfaction')

<Axes: xlabel='Gender', ylabel='count'>

More number of male and female are neutral or dissatisfied.

sns.countplot(x='customer_type',data=df,hue='satisfaction')

<Axes: xlabel='customer_type', ylabel='count'>

A large number of loyal customers are neutral or dissatisfied. Maximum disloyal customers are neutral or dissatified only.

sns.countplot(x='type_of_travel',data=df,hue='satisfaction')

<Axes: xlabel='type_of_travel', ylabel='count'>

More people who travel for personal reasons are neutral or dissatisfied. Business purpose travellers are usually satisfied. It might be because of the class they are travelling.

sns.countplot(x='customer_class',data=df,hue='type_of_travel')

<Axes: xlabel='customer_class', ylabel='count'>

Here we can see mostly people who travel for business purpose choose business class and hence are more satisfied.

sns.countplot(x='customer_class',data=df,hue='satisfaction')

<Axes: xlabel='customer_class', ylabel='count'>

People travelling with business class are usually satisfied. People travelling through Eco plus and Eco class are usually neutral or dissatisfied.

sns.countplot(x='satisfaction',data=df,hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='count'>

df['satisfaction'].value_counts()

satisfaction
neutral or dissatisfied    73452
satisfied                  56428
Name: count, dtype: int64

More number of customers are neutal or dissatisfied compared to customers who are satisfied.

sns.countplot(x='inflight_wifi_service',data=df,hue='satisfaction')

<Axes: xlabel='inflight_wifi_service', ylabel='count'>

More customers who have given 2 &3 ratings are usually dissatified. Inflight wigfi service has to be improved.

sns.countplot(x='departure_arrival_time_convenient',data=df,hue='satisfaction')

<Axes: xlabel='departure_arrival_time_convenient', ylabel='count'>

Here we can see people who have given 4 and 5 ratings are more neutral or dissatisfied. So this is not an important parameter for measuring satisfaction.

sns.countplot(x='ease_of_online_booking',data=df,hue='satisfaction')

<Axes: xlabel='ease_of_online_booking', ylabel='count'>

Customers who have given 3 & 4 ratings and neutral or dissatisfied. Only those who have given 5 ratings are satisfied. There can be some technical glitches which can be improved for those who have given 3 rating.

sns.countplot(x='gate_location',data=df,hue='satisfaction')

<Axes: xlabel='gate_location', ylabel='count'>

Here we can see people who have given 3 and 4 ratings are more neutral or dissatisfied. So this can be considered as not a very important parameter for measuring satisfaction.

sns.countplot(x='food_and_drink',data=df,hue='satisfaction')

<Axes: xlabel='food_and_drink', ylabel='count'>

Improving food and drinks can contribute to satisfaction as can be clearly seen from the chart.

sns.countplot(x='online_boarding',data=df,hue='satisfaction')

<Axes: xlabel='online_boarding', ylabel='count'>

Customer satisfaction is increasing with online boarding.

sns.countplot(x='seat_comfort',data=df,hue='satisfaction')

<Axes: xlabel='seat_comfort', ylabel='count'>

Customer satisfaction is increasing with seat comfort.

sns.countplot(x='inflight_entertainment',data=df,hue='satisfaction')

<Axes: xlabel='inflight_entertainment', ylabel='count'>

Customer satisfaction is increasing with inflight entertainment.

sns.countplot(x='onboard_service',data=df,hue='satisfaction')

<Axes: xlabel='onboard_service', ylabel='count'>

Customer satisfaction is increasing with onboard service.

sns.countplot(x='leg_room_service',data=df,hue='satisfaction')

<Axes: xlabel='leg_room_service', ylabel='count'>

Customer satisfaction is increasing with leg room service.

sns.countplot(x='baggage_handling',data=df,hue='satisfaction')

<Axes: xlabel='baggage_handling', ylabel='count'>

This plot shows slight imbalance. Satisfaction is increasing with baggage_handling. Although other parameters steer satisfaction even after 4 rating to baggage handling.

sns.countplot(x='checkin_service',data=df,hue='satisfaction')

<Axes: xlabel='checkin_service', ylabel='count'>

This plot shows that satisfaction is increasing with checkin service.

sns.countplot(x='inflight_service',data=df,hue='satisfaction')

<Axes: xlabel='inflight_service', ylabel='count'>

Satisfaction is increasing with inflight_service.

sns.countplot(x='cleanliness',data=df,hue='satisfaction')

<Axes: xlabel='cleanliness', ylabel='count'>

Satisfaction is increasing with cleanliness.

plt.subplots(figsize = (10,7))
sns.histplot(x='flight_distance',hue="satisfaction",data=df,kde=True)
plt.title("Flight Distance VS Satisfaction")

Text(0.5, 1.0, 'Flight Distance VS Satisfaction')

sns.pairplot(df)

<seaborn.axisgrid.PairGrid at 0x174565861b0>

Most people in less distance flights are neutral or dissatisfied. More focus should be given here for increasing satisfaction.

sns.boxplot(x='satisfaction',y='inflight_wifi_service',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='inflight_wifi_service'>

There are few outliers of inflight_wifi_service for neutral or dissatisfied section of satisfaction. People are usually dissatisfied when wifi service is not good.

sns.boxplot(x='satisfaction',y='age',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='age'>

The mean age of customers who are satisfied is slightly more than those who are neutral or dissatisfied.

sns.boxplot(x='satisfaction',y='departure_arrival_time_convenient',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='departure_arrival_time_convenient'>

The mean age of departure_arrival_time_convenient for both class of satisfaction are equal.

sns.boxplot(x='satisfaction',y='ease_of_online_booking',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='ease_of_online_booking'>

The mean rating of ease_of_online_booking is more for satisfied people than people who are neutral or dissatisfied. There are few outliers where people are neutral or dissatisfied.

sns.boxplot(x='satisfaction',y='gate_location',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='gate_location'>

The mean age of gate_location for both class of satisfaction are equal.

sns.boxplot(x='satisfaction',y='food_and_drink',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='food_and_drink'>

The mean rating of food_and_drink is more for satisfied people than people who are neutral or dissatisfied.

sns.boxplot(x='satisfaction',y='online_boarding',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='online_boarding'>

The mean rating of online_boarding is more for satisfied people than people who are neutral or dissatisfied. There are outliers in the data.

sns.boxplot(x='satisfaction',y='seat_comfort',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='seat_comfort'>

The mean rating of seat_comfort is more for satisfied people than people who are neutral or dissatisfied. There are outliers in the data where people are satisfied.

sns.boxplot(x='satisfaction',y='inflight_entertainment',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='inflight_entertainment'>

The mean rating of inflight_entertainment is more for satisfied people than people who are neutral or dissatisfied. There are outliers in the data where people are satisfied.

sns.boxplot(x='satisfaction',y='onboard_service',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='onboard_service'>

The mean rating of onboard_service is more for satisfied people than people who are neutral or dissatisfied.

sns.boxplot(x='satisfaction',y='leg_room_service',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='leg_room_service'>

The mean rating of leg room service is more for satisfied people than people who are neutral or dissatisfied.

sns.boxplot(x='satisfaction',y='baggage_handling',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='baggage_handling'>

The mean rating of baggage handling is more for satisfied people than people who are neutral or dissatisfied. There are outliers in the data.

sns.boxplot(x='satisfaction',y='checkin_service',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='checkin_service'>

The mean rating of checkin service is more for satisfied people than people who are neutral or dissatisfied.

sns.boxplot(x='satisfaction',y='inflight_service',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='inflight_service'>

The mean rating inflight service is more for satisfied people than people who are neutral or dissatisfied. There are outliers in the data.

sns.boxplot(x='satisfaction',y='cleanliness',data=df, hue='satisfaction')

<Axes: xlabel='satisfaction', ylabel='cleanliness'>

The mean rating of cleanliness is more for satisfied people than people who are neutral or dissatisfied.

import seaborn as sns
sns.boxplot(x='departure_delay_in_minutes', figsize =(10,6), data=df)
plt.show()

sns.boxplot(x='arrival_delay_in_minutes', data=df)
plt.show()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[3], line 1
----> 1 sns.boxplot(x='arrival_delay_in_minutes', data=df)
      2 plt.show()

NameError: name 'df' is not defined

def remove_outliers(column_name):
    Q1 = df[column_name].quantile(0.25)

    Q3 = df[column_name].quantile(0.75)

    IQR = Q3 - Q1

    Upper_boundary = Q3 + (1.5 * IQR)

    Lower_boundary = Q1 - (1.5 * IQR)

    # replace number > Upper_boundary and number < Lower_boundary with nan value
    df.loc[(df[column_name] > Upper_boundary) | (df[column_name] < Lower_boundary), column_name] = np.nan

    # replace nan value with mean
    df[column_name].fillna(df[column_name].mode(), inplace = True)

remove_outliers("inflight_service")
remove_outliers("baggage_handling")
remove_outliers("inflight_entertainment")
remove_outliers("seat_comfort")
remove_outliers("online_boarding")
remove_outliers("checkin_service")

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3950046972.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df[column_name].fillna(df[column_name].mode(), inplace = True)
C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3950046972.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df[column_name].fillna(df[column_name].mode(), inplace = True)
C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3950046972.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df[column_name].fillna(df[column_name].mode(), inplace = True)
C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3950046972.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df[column_name].fillna(df[column_name].mode(), inplace = True)
C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3950046972.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df[column_name].fillna(df[column_name].mode(), inplace = True)
C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\3950046972.py:16: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  df[column_name].fillna(df[column_name].mode(), inplace = True)

# Ensure columns exist and then encode them
columns_to_encode = ['Gender', 'customer_type', 'type_of_travel', 'customer_class', 'satisfaction']

for column in columns_to_encode:
    if column in df.columns:
        df[column] = df[column].astype('category').cat.codes
    else:
        print(f"Column '{column}' not found in the DataFrame")

# Verify the changes
print(df.head())

   Gender  customer_type  age  type_of_travel  customer_class  \
0       1              0   13               1               2   
1       1              1   25               0               0   
2       0              0   26               0               0   
3       0              0   25               0               0   
4       1              0   61               0               0   

   flight_distance  inflight_wifi_service  departure_arrival_time_convenient  \
0              460                      3                                  4   
1              235                      3                                  2   
2             1142                      2                                  2   
3              562                      2                                  5   
4              214                      3                                  3   

   ease_of_online_booking  gate_location  ...  inflight_entertainment  \
0                       3              1  ...                     5.0   
1                       3              3  ...                     1.0   
2                       2              2  ...                     5.0   
3                       5              5  ...                     2.0   
4                       3              3  ...                     3.0   

   onboard_service  leg_room_service  baggage_handling  checkin_service  \
0                4                 3               4.0              4.0   
1                1                 5               3.0              NaN   
2                4                 3               4.0              4.0   
3                2                 5               3.0              NaN   
4                3                 4               4.0              3.0   

   inflight_service  cleanliness  departure_delay_in_minutes  \
0               5.0            5                          25   
1               4.0            1                           1   
2               4.0            5                           0   
3               4.0            2                          11   
4               3.0            3                           0   

   arrival_delay_in_minutes  satisfaction  
0                      18.0             0  
1                       6.0             0  
2                       0.0             1  
3                       9.0             0  
4                       0.0             1  

[5 rows x 23 columns]

corr = df.corr()

corr

	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	gate_location	...	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	arrival_delay_in_minutes	satisfaction
Gender	1.000000	-0.030958	0.008996	0.009503	-0.011574	0.003616	0.005901	0.008995	0.005893	-0.000863	...	0.003843	0.006447	0.031047	0.036356	0.010361	0.038504	0.002867	0.003491	0.001309	0.011236
customer_type	-0.030958	1.000000	-0.284172	-0.308236	0.042994	-0.226021	-0.005757	-0.206873	-0.018059	0.004465	...	-0.106001	-0.054172	-0.046841	0.024874	-0.024412	0.023292	-0.081302	0.003859	0.004730	-0.186017
age	0.008996	-0.284172	1.000000	-0.044808	-0.116921	0.099459	0.016116	0.036960	0.022565	-0.000398	...	0.074947	0.057078	0.039119	-0.047991	0.023708	-0.051347	0.052565	-0.009041	-0.011248	0.134091
type_of_travel	0.009503	-0.308236	-0.044808	1.000000	0.486718	-0.266792	-0.105865	0.257102	-0.134078	-0.029869	...	-0.152936	-0.059794	-0.139612	-0.033012	0.012815	-0.023538	-0.084615	-0.005913	-0.005830	-0.449861
customer_class	-0.011574	0.042994	-0.116921	0.486718	1.000000	-0.426925	-0.024962	0.087152	-0.095138	-0.005701	...	-0.183162	-0.210649	-0.198825	-0.166588	-0.118912	-0.159285	-0.129623	0.009530	0.014162	-0.448193
flight_distance	0.003616	-0.226021	0.099459	-0.266792	-0.426925	1.000000	0.006701	-0.018914	0.065165	0.005520	...	0.130507	0.111194	0.134533	0.064855	0.055870	0.059316	0.095648	0.002402	-0.001935	0.298085
inflight_wifi_service	0.005901	-0.005757	0.016116	-0.105865	-0.024962	0.006701	1.000000	0.344915	0.714807	0.338573	...	0.207802	0.119928	0.160317	0.120376	0.035005	0.110029	0.131300	-0.015946	-0.017749	0.283460
departure_arrival_time_convenient	0.008995	-0.206873	0.036960	0.257102	0.087152	-0.018914	0.344915	1.000000	0.437620	0.447510	...	-0.008380	0.067297	0.010617	0.070833	0.071985	0.072195	0.009862	0.000778	-0.000942	-0.054270
ease_of_online_booking	0.005893	-0.018059	0.022565	-0.134078	-0.095138	0.065165	0.714807	0.437620	1.000000	0.460041	...	0.046564	0.039064	0.109450	0.039148	0.002264	0.035373	0.015125	-0.005318	-0.007033	0.168877
gate_location	-0.000863	0.004465	-0.000398	-0.029869	-0.005701	0.005520	0.338573	0.447510	0.460041	1.000000	...	0.002741	-0.029019	-0.005181	0.000972	-0.034097	0.000310	-0.005918	0.005973	0.005658	-0.002793
food_and_drink	0.001730	-0.056997	0.023194	-0.068986	-0.080773	0.057066	0.132214	0.000687	0.030514	-0.002872	...	0.623461	0.057404	0.033173	0.035321	0.070585	0.035210	0.658054	-0.029164	-0.031715	0.211340
online_boarding	-0.045022	-0.189083	0.207572	-0.224020	-0.297645	0.214825	0.457445	0.072287	0.404866	0.002756	...	0.283922	0.154242	0.123225	0.083541	0.163077	0.074058	0.329377	-0.019404	-0.022730	0.501749
seat_comfort	-0.030756	-0.156239	0.159136	-0.127717	-0.212334	0.157662	0.121513	0.008666	0.028561	0.002788	...	0.611837	0.130545	0.104272	0.074620	0.153768	0.068842	0.679613	-0.027999	-0.030521	0.348829
inflight_entertainment	0.003843	-0.106001	0.074947	-0.152936	-0.183162	0.130507	0.207802	-0.008380	0.046564	0.002741	...	1.000000	0.418574	0.300397	0.379123	0.101803	0.406094	0.692511	-0.027012	-0.030230	0.398234
onboard_service	0.006447	-0.054172	0.057078	-0.059794	-0.210649	0.111194	0.119928	0.067297	0.039064	-0.029019	...	0.418574	1.000000	0.357721	0.520296	0.201324	0.551569	0.122084	-0.030486	-0.034789	0.322205
leg_room_service	0.031047	-0.046841	0.039119	-0.139612	-0.198825	0.134533	0.160317	0.010617	0.109450	-0.005181	...	0.300397	0.357721	1.000000	0.371455	0.126902	0.369569	0.096695	0.014574	0.011346	0.312424
baggage_handling	0.036356	0.024874	-0.047991	-0.033012	-0.166588	0.064855	0.120376	0.070833	0.039148	0.000972	...	0.379123	0.520296	0.371455	1.000000	0.188843	0.629237	0.097071	-0.004105	-0.007935	0.248680
checkin_service	0.010361	-0.024412	0.023708	0.012815	-0.118912	0.055870	0.035005	0.071985	0.002264	-0.034097	...	0.101803	0.201324	0.126902	0.188843	1.000000	0.190977	0.146582	-0.013459	-0.016034	0.198172
inflight_service	0.038504	0.023292	-0.051347	-0.023538	-0.159285	0.059316	0.110029	0.072195	0.035373	0.000310	...	0.406094	0.551569	0.369569	0.629237	0.190977	1.000000	0.090356	-0.054432	-0.059853	0.244918
cleanliness	0.002867	-0.081302	0.052565	-0.084615	-0.129623	0.095648	0.131300	0.009862	0.015125	-0.005918	...	0.692511	0.122084	0.096695	0.097071	0.146582	0.090356	1.000000	-0.014543	-0.016546	0.307035
departure_delay_in_minutes	0.003491	0.003859	-0.009041	-0.005913	0.009530	0.002402	-0.015946	0.000778	-0.005318	0.005973	...	-0.027012	-0.030486	0.014574	-0.004105	-0.013459	-0.054432	-0.014543	1.000000	0.965291	-0.050740
arrival_delay_in_minutes	0.001309	0.004730	-0.011248	-0.005830	0.014162	-0.001935	-0.017749	-0.000942	-0.007033	0.005658	...	-0.030230	-0.034789	0.011346	-0.007935	-0.016034	-0.059853	-0.016546	0.965291	1.000000	-0.058275
satisfaction	0.011236	-0.186017	0.134091	-0.449861	-0.448193	0.298085	0.283460	-0.054270	0.168877	-0.002793	...	0.398234	0.322205	0.312424	0.248680	0.198172	0.244918	0.307035	-0.050740	-0.058275	1.000000

23 rows × 23 columns

plt.figure(figsize=(30,5))
sns.heatmap(df.select_dtypes(exclude = object).corr(), fmt = ".2f", linewidths = 0.2, annot=True,cmap='coolwarm',center=0)
plt.show()

MODELLING

PREPROCESSING FOR THE MODEL We want to predict the passenger satisfaction.

#importing required models
from statsmodels.stats.outliers_influence import variance_inflation_factor
import statsmodels.api as sm
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import accuracy_score, confusion_matrix, classification_report
from sklearn.ensemble import RandomForestRegressor

#checking multicollinearity

import statsmodels.api as sm
from statsmodels.stats.outliers_influence import variance_inflation_factor


# Separate features and target variable
x = df.drop(columns=['satisfaction'])
y = df['satisfaction'].values

# Check for missing or infinite values in the features
if x.isnull().values.any() or np.isinf(x).values.any():

    # Option 2: Impute missing values (example: fill with mean)
     x = x.fillna(x.mean()).replace([np.inf, -np.inf], np.nan).fillna(x.mean())

# Add a constant to the independent variables
X = sm.add_constant(x)

# Calculate VIF for each independent variable
vif_data = pd.DataFrame()
vif_data["feature"] = X.columns
vif_data["VIF"] = [variance_inflation_factor(X.values, i) for i in range(X.shape[1])]

print(vif_data)

                              feature        VIF
0                               const  60.406934
1                              Gender   1.011189
2                       customer_type   1.514992
3                                 age   1.161430
4                      type_of_travel   1.831717
5                      customer_class   1.680157
6                     flight_distance   1.344833
7               inflight_wifi_service   2.442128
8   departure_arrival_time_convenient   1.663603
9              ease_of_online_booking   2.694192
10                      gate_location   1.507225
11                     food_and_drink   2.173924
12                    online_boarding   1.983909
13                       seat_comfort   2.382917
14             inflight_entertainment   3.814593
15                    onboard_service   1.759608
16                   leg_room_service   1.315638
17                   baggage_handling   1.900141
18                    checkin_service   1.112991
19                   inflight_service   2.064726
20                        cleanliness   2.846353
21         departure_delay_in_minutes  12.867672
22           arrival_delay_in_minutes  12.878252

df= df.drop(columns=['arrival_delay_in_minutes'],axis=1)
df

	Gender	customer_type	age	type_of_travel	customer_class	flight_distance	inflight_wifi_service	departure_arrival_time_convenient	ease_of_online_booking	gate_location	...	seat_comfort	inflight_entertainment	onboard_service	leg_room_service	baggage_handling	checkin_service	inflight_service	cleanliness	departure_delay_in_minutes	satisfaction
0	1	0	13	1	2	460	3	4	3	1	...	5.0	5.0	4	3	4.0	4.0	5.0	5	25	0
1	1	1	25	0	0	235	3	2	3	3	...	1.0	1.0	1	5	3.0	NaN	4.0	1	1	0
2	0	0	26	0	0	1142	2	2	2	2	...	5.0	5.0	4	3	4.0	4.0	4.0	5	0	1
3	0	0	25	0	0	562	2	5	5	5	...	2.0	2.0	2	5	3.0	NaN	4.0	2	11	0
4	1	0	61	0	0	214	3	3	3	3	...	5.0	3.0	3	4	4.0	3.0	3.0	3	0	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
129875	1	1	34	0	0	526	3	3	3	1	...	4.0	4.0	3	2	4.0	4.0	5.0	4	0	0
129876	1	0	23	0	0	646	4	4	4	4	...	4.0	4.0	4	5	5.0	5.0	5.0	4	0	1
129877	0	0	17	1	1	828	2	5	1	5	...	2.0	2.0	4	3	4.0	5.0	4.0	2	0	0
129878	1	0	14	0	0	1127	3	3	3	3	...	4.0	4.0	3	2	5.0	4.0	5.0	4	0	1
129879	0	0	42	1	1	264	2	5	2	5	...	2.0	1.0	1	2	1.0	NaN	1.0	1	0	0

129880 rows × 22 columns

df.shape

(129880, 22)

#Chceking VIF again

# Separate features and target variable
x = df.drop(columns=['satisfaction'])
y = df['satisfaction'].values

# Check for missing or infinite values in the features
if x.isnull().values.any() or np.isinf(x).values.any():
    # Option 1: Drop rows with missing or infinite values
    #x = x.dropna().replace([np.inf, -np.inf], np.nan).dropna()

    # Option 2: Impute missing values (example: fill with mean)
     x = x.fillna(x.mean()).replace([np.inf, -np.inf], np.nan).fillna(x.mean())

# Add a constant to the independent variables
X = sm.add_constant(x)

# Calculate VIF for each independent variable
vif_data = pd.DataFrame()
vif_data["feature"] = X.columns
vif_data["VIF"] = [variance_inflation_factor(X.values, i) for i in range(X.shape[1])]

print(vif_data)

                              feature        VIF
0                               const  60.346700
1                              Gender   1.011144
2                       customer_type   1.514927
3                                 age   1.161353
4                      type_of_travel   1.831506
5                      customer_class   1.679981
6                     flight_distance   1.344703
7               inflight_wifi_service   2.442117
8   departure_arrival_time_convenient   1.663601
9              ease_of_online_booking   2.694190
10                      gate_location   1.507222
11                     food_and_drink   2.173681
12                    online_boarding   1.983884
13                       seat_comfort   2.382893
14             inflight_entertainment   3.814509
15                    onboard_service   1.759570
16                   leg_room_service   1.315632
17                   baggage_handling   1.900097
18                    checkin_service   1.112987
19                   inflight_service   2.063934
20                        cleanliness   2.846334
21         departure_delay_in_minutes   1.007577

# Split the dataset into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

Feature Scaling

# Standardize the features
scaler = StandardScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)

ML MODELS BEFORE RESAMPLING

# Train the logistic regression model
model = LogisticRegression()
model.fit(X_train_scaled, y_train)

LogisticRegression()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

# Make predictions
y_pred = model.predict(X_test_scaled)

# Evaluate the model
print("Confusion Matrix:")
print(confusion_matrix(y_test, y_pred))
print("\nClassification Report:")
print(classification_report(y_test, y_pred))

Confusion Matrix:
[[13200  1422]
 [ 1864  9490]]

Classification Report:
              precision    recall  f1-score   support

           0       0.88      0.90      0.89     14622
           1       0.87      0.84      0.85     11354

    accuracy                           0.87     25976
   macro avg       0.87      0.87      0.87     25976
weighted avg       0.87      0.87      0.87     25976

from sklearn.neighbors import KNeighborsClassifier

knn= KNeighborsClassifier()

knn.fit(X_train_scaled,y_train)

KNeighborsClassifier()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

y_pred_knn=knn.predict(X_test_scaled)

from sklearn.metrics import accuracy_score 
accuracy_score(y_test, y_pred_knn)

0.9291268863566369

from sklearn.metrics import classification_report
print(classification_report(y_test,y_pred_knn))

              precision    recall  f1-score   support

           0       0.91      0.97      0.94     14622
           1       0.95      0.88      0.92     11354

    accuracy                           0.93     25976
   macro avg       0.93      0.92      0.93     25976
weighted avg       0.93      0.93      0.93     25976

from sklearn.tree import DecisionTreeClassifier

dt= DecisionTreeClassifier()
dt.fit(X_train_scaled,y_train)

DecisionTreeClassifier()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

y_pred_dt=dt.predict(X_test_scaled)
print(accuracy_score(y_test, y_pred_dt))

0.9455651370495842

print(classification_report(y_test,y_pred_dt))

              precision    recall  f1-score   support

           0       0.95      0.95      0.95     14622
           1       0.94      0.94      0.94     11354

    accuracy                           0.95     25976
   macro avg       0.94      0.94      0.94     25976
weighted avg       0.95      0.95      0.95     25976

from sklearn.ensemble import RandomForestClassifier
ran_forest=RandomForestClassifier()
ran_forest.fit(X_train_scaled,y_train)

RandomForestClassifier()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

y_pred_forest=ran_forest.predict(X_test_scaled)
print(accuracy_score(y_test,y_pred_forest))

0.962157376039421

print(classification_report(y_test,y_pred_forest))

              precision    recall  f1-score   support

           0       0.95      0.98      0.97     14622
           1       0.98      0.94      0.96     11354

    accuracy                           0.96     25976
   macro avg       0.96      0.96      0.96     25976
weighted avg       0.96      0.96      0.96     25976

from sklearn.svm import SVC
svc=SVC()
svc.fit(X_train_scaled,y_train)

SVC()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

y_pred_svc=svc.predict(X_test_scaled)
print(accuracy_score(y_test,y_pred_svc))

0.9541499846011703

print(classification_report(y_test,y_pred_svc))

              precision    recall  f1-score   support

           0       0.95      0.97      0.96     14622
           1       0.96      0.93      0.95     11354

    accuracy                           0.95     25976
   macro avg       0.95      0.95      0.95     25976
weighted avg       0.95      0.95      0.95     25976

ML MODELS AFTER RESAMPLING

pip install imblearn

Requirement already satisfied: imblearn in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (0.0)Note: you may need to restart the kernel to use updated packages.

Requirement already satisfied: imbalanced-learn in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (from imblearn) (0.12.3)
Requirement already satisfied: numpy>=1.17.3 in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (from imbalanced-learn->imblearn) (2.0.0)
Requirement already satisfied: scipy>=1.5.0 in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (from imbalanced-learn->imblearn) (1.14.0)
Requirement already satisfied: scikit-learn>=1.0.2 in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (from imbalanced-learn->imblearn) (1.5.1)
Requirement already satisfied: joblib>=1.1.1 in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (from imbalanced-learn->imblearn) (1.4.2)
Requirement already satisfied: threadpoolctl>=2.0.0 in c:\users\ramasish chatterjee\appdata\local\programs\python\python312\lib\site-packages (from imbalanced-learn->imblearn) (3.5.0)

from imblearn.over_sampling import SMOTE

sm=SMOTE(random_state=42)
x_res,y_res=sm.fit_resample(X_train_scaled,y_train)

knn.fit(x_res,y_res)
y_pred_knn_res=knn.predict(X_test_scaled) 
accuracy_score(y_test, y_pred_knn_res)
print(classification_report(y_test,y_pred_knn_res))

              precision    recall  f1-score   support

           0       0.92      0.95      0.94     14622
           1       0.94      0.90      0.92     11354

    accuracy                           0.93     25976
   macro avg       0.93      0.93      0.93     25976
weighted avg       0.93      0.93      0.93     25976

dt.fit(x_res,y_res)
y_pred_dt_res=dt.predict(X_test_scaled) 
accuracy_score(y_test, y_pred_dt_res)
print(classification_report(y_test,y_pred_dt_res))

              precision    recall  f1-score   support

           0       0.95      0.95      0.95     14622
           1       0.94      0.94      0.94     11354

    accuracy                           0.94     25976
   macro avg       0.94      0.94      0.94     25976
weighted avg       0.94      0.94      0.94     25976

ran_forest.fit(x_res,y_res)
y_pred_ran_forest_res=ran_forest.predict(X_test_scaled) 
accuracy_score(y_test, y_pred_ran_forest_res)
print(classification_report(y_test,y_pred_ran_forest_res))

              precision    recall  f1-score   support

           0       0.95      0.98      0.97     14622
           1       0.97      0.94      0.96     11354

    accuracy                           0.96     25976
   macro avg       0.96      0.96      0.96     25976
weighted avg       0.96      0.96      0.96     25976

svc.fit(x_res,y_res)
y_pred_svc_res=svc.predict(X_test_scaled) 
accuracy_score(y_test, y_pred_svc_res)
print(classification_report(y_test,y_pred_svc_res))

              precision    recall  f1-score   support

           0       0.95      0.96      0.96     14622
           1       0.95      0.94      0.95     11354

    accuracy                           0.95     25976
   macro avg       0.95      0.95      0.95     25976
weighted avg       0.95      0.95      0.95     25976

HYPERPARAMETER TUNING

from sklearn.model_selection import GridSearchCV

KNN

knn_params={"n_neighbors":[1,10,30,50],"weights":["uniform","distance"]}
knn_grid=GridSearchCV(knn,knn_params,cv=5)
knn_grid.fit(x_res,y_res)

C:\Users\Ramasish Chatterjee\AppData\Local\Programs\Python\Python312\Lib\site-packages\numpy\ma\core.py:2846: RuntimeWarning: invalid value encountered in cast
  _data = np.array(data, dtype=dtype, copy=copy,

GridSearchCV(cv=5, estimator=KNeighborsClassifier(),
             param_grid={'n_neighbors': [1, 10, 30, 50],
                         'weights': ['uniform', 'distance']})

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

knn_result=pd.DataFrame(knn_grid.cv_results_)
best_parameters_knn=knn_grid.best_params_
print(best_parameters_knn,knn_grid.best_score_)

{'n_neighbors': 10, 'weights': 'distance'} 0.9416624171341151

Decision Tree

dt_params={'criterion':['gini','entropy'],'max_depth':[1,3,5,7,10,None]}
dt_grid=GridSearchCV(dt,dt_params,cv=5)
dt_grid.fit(x_res,y_res)

GridSearchCV(cv=5, estimator=DecisionTreeClassifier(),
             param_grid={'criterion': ['gini', 'entropy'],
                         'max_depth': [1, 3, 5, 7, 10, None]})

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

dt_result=pd.DataFrame(dt_grid.cv_results_)
best_parameters_dt=dt_grid.best_params_
print(best_parameters_dt,dt_grid.best_score_)

{'criterion': 'entropy', 'max_depth': None} 0.9519377868434471

Random Forest

ran_forest_params={"criterion":['gini','entropy'],"max_depth":[1,2,5,7,None]}
ran_forest_grid=GridSearchCV(ran_forest,ran_forest_params,cv=5)
ran_forest_grid.fit(x_res,y_res)

GridSearchCV(cv=5, estimator=RandomForestClassifier(),
             param_grid={'criterion': ['gini', 'entropy'],
                         'max_depth': [1, 2, 5, 7, None]})

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

ran_forest_result=pd.DataFrame(ran_forest_grid.cv_results_)
best_parameters_ran_forest=ran_forest_grid.best_params_
print(best_parameters_ran_forest,ran_forest_grid.best_score_)

{'criterion': 'gini', 'max_depth': None} 0.9662417134115246

FINAL MODELS

from sklearn.metrics import roc_curve

knn1= KNeighborsClassifier(n_neighbors=10,weights='distance')

clf1=knn1.fit(x_res,y_res)
y_pred_knn1=knn1.predict(X_test_scaled)

accuracy_score(y_test, y_pred_knn1)
print(classification_report(y_test,y_pred_knn1))

              precision    recall  f1-score   support

           0       0.92      0.96      0.94     14622
           1       0.94      0.90      0.92     11354

    accuracy                           0.93     25976
   macro avg       0.93      0.93      0.93     25976
weighted avg       0.93      0.93      0.93     25976

dtf=DecisionTreeClassifier(criterion="entropy", max_depth=None)

dtf.fit(x_res,y_res)
y_pred_dtf=dtf.predict(X_test_scaled) 
accuracy_score(y_test, y_pred_dtf)
print(classification_report(y_test,y_pred_dtf))

              precision    recall  f1-score   support

           0       0.95      0.95      0.95     14622
           1       0.94      0.94      0.94     11354

    accuracy                           0.95     25976
   macro avg       0.95      0.95      0.95     25976
weighted avg       0.95      0.95      0.95     25976

ran_forestf=RandomForestClassifier(criterion="entropy",max_depth=None)
ran_forestf.fit(x_res,y_res)
y_pred_ran_forest_f=ran_forest.predict(X_test_scaled) 
accuracy_score(y_test, y_pred_ran_forest_f)
print(classification_report(y_test,y_pred_ran_forest_f))

              precision    recall  f1-score   support

           0       0.95      0.98      0.97     14622
           1       0.97      0.94      0.96     11354

    accuracy                           0.96     25976
   macro avg       0.96      0.96      0.96     25976
weighted avg       0.96      0.96      0.96     25976

Feature Importance

# Get the feature importances
importances = ran_forestf.feature_importances_

# Convert X_train_scaled back to DataFrame to get column names
X_train_scaled_df = pd.DataFrame(X_train_scaled, columns=X_train.columns)

# Map the feature importances to their corresponding feature names
feature_importance_df = pd.DataFrame({
    'Feature': X_train_scaled_df.columns,
    'Importance': importances
}).sort_values(by='Importance', ascending=False)

# Display the feature importances
print(feature_importance_df)

# Plot the feature importances
plt.figure(figsize=(10, 8))
sns.barplot(x='Importance', y='Feature', data=feature_importance_df, palette='viridis')
plt.title('Feature Importances from Random Forest Classifier')
plt.xlabel('Importance')
plt.ylabel('Feature')
plt.show()

                              Feature  Importance
7               inflight_wifi_service    0.159751
12                    online_boarding    0.150886
4                      type_of_travel    0.094083
5                      customer_class    0.089571
14             inflight_entertainment    0.050554
13                       seat_comfort    0.046087
9              ease_of_online_booking    0.046067
16                   leg_room_service    0.041740
2                       customer_type    0.040793
6                     flight_distance    0.038738
3                                 age    0.037743
15                    onboard_service    0.030564
17                   baggage_handling    0.030010
20                        cleanliness    0.026325
19                   inflight_service    0.026037
18                    checkin_service    0.022429
10                      gate_location    0.018200
8   departure_arrival_time_convenient    0.017686
21         departure_delay_in_minutes    0.014955
11                     food_and_drink    0.012665
1                              Gender    0.005117
0                               const    0.000000

C:\Users\Ramasish Chatterjee\AppData\Local\Temp\ipykernel_1784\1183606931.py:18: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect.

  sns.barplot(x='Importance', y='Feature', data=feature_importance_df, palette='viridis')

df.shape

(129880, 22)